Drug Interactions Hormonal

BackgroundThis Guidance updates a Faculty Aid to ContinuingProfessional Development Topic (FACT) on druginteractions with hormonal contraception.1 This Guidancedoes not consider the effects on hormonal contraception ofthe underlying condition that necessitated concurrentmedication.

Potential drug interactions should be considered whenprescribing any medication for women of reproductive age.Concentrations of contraceptive hormones may beincreased or decreased by concomitant drug use.Contraceptive hormones may themselves increase ordecrease serum concentrations of concomitant drugs. Druginteractions that result in a reduction in contraceptiveefficacy are clinically relevant. Factors that influence theclinical significance of interactions include: narrowtherapeutic ranges of drugs or hormones, wide individualvariation in serum concentrations of drugs or hormones,and poor compliance.

There is a lack of good quality, robust evidence on theeffects of drugs on hormonal contraception. Most data havebeen obtained from case reports, which provide limitedevidence. Pregnancy has been reported in women usinghormonal contraception following use of concomitantdrugs.2–6 Nevertheless, this does not prove that the drugwas responsible for contraceptive failure leading topregnancy.

The British National Formulary (BNF) providesclinicians with information on potential druginteractions. Nevertheless, the BNF uses informationfrom Summaries of Product Characteristics (SPCs),which may not reflect current evidence. This evidence-based Guidance summarises evidence on interactionsbetween hormonal contraception and liver enzyme-inducing drugs, non-liver enzyme-inducing antibiotics,drugs which may be toxic if serum concentrationsincrease, and on commonly used drugs (prescription andnon-prescription) which do and do not affectcontraceptive efficacy.

Further research into drug interactions and hormonalcontraception is needed to guide clinical practice andshould be encouraged.

How might concomitant drugs affect contraceptivehormones?Drug interactions may result from alterations inpharmacodynamics or pharmacokinetics.7

Pharmacodynamic interactions occur when one drugdirectly influences the action of another by synergy orantagonism. There are no important pharmacodynamicinteractions relevant for hormonal contraception.

Pharmacokinetic interactions occur during theprocesses of drug absorption, distribution, metabolism orelimination. Pharmacokinetic interactions are relevant tocontraceptive hormones (as summarised in Figure 1).

Bioavailibility is the amount of hormone available tohave clinical effect (e.g. inhibition of ovulation or cervicalmucus thickening). There are wide inter- and intra-individual variations in the bioavailability ofethinylestradiol (EE) and progestogens.8 Bioavailability ofcontraceptive hormones depends primarily on absorption(including secondary absorption via the enterohepaticcirculation) and metabolism.

AbsorptionOrally administered EE and progestogens are absorbedfrom the small intestine. Drugs that affect the absorption ofhormones may theoretically reduce bioavailability but noevidence was identified to support this. Drugs that inducevomiting may indirectly affect hormone absorption. Ifvomiting occurs less than 2 hours after ingesting hormonalcontraception, women are advised to take a furtherdose.9–11 It is unclear whether crushing or chewingcontraceptive pills can affect absorption. A chewablecombined oral contraceptive (COC) is available in theUSA. This should be taken with a glass of water to ensureall of the dose reaches the stomach.12

Both EE and progestogens can be absorbed directly intothe systemic circulation (thus bypassing first-passmetabolism in the intestinal mucosa and liver).Progestogens and EE are absorbed transdermally via thecombined contraceptive patch13 or vaginally via acontraceptive ring.14

When progestogen is administered as an etonorgestrel(ENG) subdermal implant, almost 100% is bioavailable.

139J Fam Plann Reprod Health Care 2005: 31(2)

This Guidance provides information for clinicians and women using hormonal contraception applicable when concurrentmedications are prescribed. A key to the grades of recommendations, based on levels of evidence, is given at the end of thisdocument. Details of the methods used by the Clinical Effectiveness Unit (CEU) in developing this Guidance and evidencetables summarising the research basis of the recommendations are available on the Faculty website (www.ffprhc.org.uk).Abbreviations (in alphabetical order) used include: CEU, Clinical Effectiveness Unit; COC, combined oralcontraceptive/contraception; DMPA, depot medroxyprogesterone acetate; EE, ethinylestradiol; ENG, etonorgestrel; IUD,copper-bearing intrauterine contraceptive device; LNG, levonorgestrel; LNG-IUS, levonorgestrel-releasing intrauterinesystem; POC, progestogen-only contraceptive/contraception; POP, progestogen-only pill; SJW, St John’s Wort; WHO,World Health Organization; WHOMEC, WHO Medical Eligibility Criteria for Contraceptive Use; WHOSPR, WHOSelected Practice Recommendations for Contraceptive Use.

FFPRHC Guidance (April 2005)Drug interactions with hormonal contraception

Journal of Family Planning and Reproductive Health Care 2005; 31(2): 139–151

Faculty of Family Planning and Reproductive Health CareClinical Effectiveness UnitA unit funded by the FFPRHC and supported by the University of Aberdeen and the ScottishProgramme for Clinical Effectiveness in Reproductive Health (SPCERH) to provide guidanceon evidence-based practice

ARCHIV

ED

Serum ENG concentrations increase rapidly within 8 hoursof insertion and peak after 4 days.15,16 Levels reach steadystate (200 pg/ml) after 4–6 months. These lowconcentrations are sufficient to inhibit ovulation for 3years.16

Following a single intramuscular injection ofmedroxyprogesterone acetate (MPA), progestogen isabsorbed from its injection site and metabolised in the liver.Serum concentrations of MPA are relatively constant for2–3 months (1 ng/ml).17 Repeat injections at 12-weeklyintervals maintain serum concentrations of MPA. If notrepeated, serum concentrations gradually decline and by 6months reach concentrations (0.2 ng/ml) that cannot inhibitovulation. MPA has been identified in serum 9 months aftera single injection.18

Following a single intramuscular injection ofnorethisterone enanthate, serum concentrations ofnorethisterone and norethisterone enanthate peak at 7days19 and the injection is repeated after 8 weeks tomaintain serum concentrations. Anovulation has beendemonstrated even after 12 weeks.20

DistributionOnce absorbed, EE and progestogens are transported to theliver in blood, bound to albumin or sex hormone-bindingglobulin, or free (unbound).7

MetabolismOrally administered EE and progestogens undergoextensive ‘first-pass metabolism’ in the small intestinalmucosa and liver before reaching the systemiccirculation.21,22 As much as 60% of orally administeredEE undergoes first-pass metabolism and thus only 40% isbioavailable. The bioavailibility of progestogens varies.Some hormones are ingested in inactive forms (prodrugs)and are active only after metabolism. For example,mestranol is metabolised to EE; desogestrel is metabolisedto 3-ENG; norgestimate is in part metabolised tolevonorgestrel; and ethynodiol diacetate is metabolised tonorethisterone.1

EE is metabolised in the mucosa of the small intestineand in the liver, forming sulphate and glucuronideconjugates (Figure 1).

140 J Fam Plann Reprod Health Care 2005: 31(2)

CEU GUIDANCE

Table 1 Drugs that induce liver enzymes and relevant associated drugs that do not induce liver enzymes

Type of drug Liver enzyme Effect No liver enzyme Effectinduction induction

Anti-epileptic Carbamazepine Reduction in ethinylestradiol (EE) and Ethosuximide No reduction in EE orOxcarbazepine progestogens28–31 Gabapentin progestogens32–36

Phenytoin LamotriginePhenobarbital LevetiracetamPrimidone ValproateTopiramate Vigabatrin

Antibiotic Rifampicin Reduction in EE and progestogen, breakthrough All other antibiotics Potential reduction inRifabutin bleeding27,58,59 narrow- and broad- EE due to effect on

spectrum gut flora

Antifungal Griseofulvin Known to be a potent liver enzyme-inducer, Fluconazole No alterations in EE orpregnancies documented45 progestogens55,56

Itraconazole Breakthrough bleeding, Ketoconazole no ovulation51–54

Antiretroviral Protease inhibitors Protease inhibitors

Amprenavir Reduction in EE and progestogen but additional Indinavir No clinically significant Atazanavir or alternative contraceptive methods advised interaction79

Nelfinavir with hormonal contraception79

LopinavirSaquinavir

Ritonavir EE reduced80

Non-nucleoside Nucleoside reverse reverse transcriptase transcriptaseinhibitors inhibitors

Efavirenz Reduction in EE and progestogen. Additional Abacavir No evidence forNevirapine or alternative contraceptive methods advised Didanosine pharmacokinetic

with hormonal contraception79 Emtricitabine interactions withLamivudine EE and progestogensStavudine identifiedTenofovirZalcitabineZidovudine

Gastrointestinal Lansoprazole Can induce liver enzymes but no reduction in EE 83,84

Immunosuppressant Tacrolimus Can induce liver enzymes but no published evidenceof reduced contraceptive efficacy

Respiratory Bosentan Induces liver enzymes but no evidence published Anti-asthmatic drugs No effect on hormonalfor efficacy of hormonal contraception contraception

Central nervous system Modafinil Known liver enzyme-inducer27,86 Selective serotonin No effect on contraceptivereuptake inhibitors efficacy suspected

The degree of reabsorption of EE via the enterohepaticcirculation may vary between individuals. The importanceof this reabsorption of EE, in terms of contraceptiveefficacy, is unclear. For some women it may be important,especially if contraceptive pills are missed. Women whohave had a colectomy and ileostomy have no enterohepaticcirculation of EE. The efficacy of combined oralcontraception (COC) does not appear to be reduced in thissituation.24

There is no enterohepatic circulation of progestogensand therefore antibiotics that eliminate colonic bacteria donot affect progestogen-only contraceptives (POCs).

What should be discussed when prescribing drugsto women using hormonal contraception?

1 Clinicians should consider the possibility of a druginteraction when prescribing contraception andwhen prescribing other medicines to women usinghormonal contraception (Good Practice Point).

2 Clinicians giving women information oncontraceptive options should enquire about:current and previous drug use; prescription, non-prescription and herbal drug use; and specificallyabout use of drugs which induce liver enzymes andnon-liver enzyme-inducing antibiotics (GoodPractice Point).

3 Women should be informed that some drugs mightreduce the effectiveness of hormonal contraceptionand should be advised where to seek advice if otherdrugs are taken (Good Practice Point).

4 After counselling, women using short courses ofdrugs that interact with hormonal contraceptionmay choose to continue their current hormonalmethod even if additional contraception, such ascondoms, is required. However, women on long-term courses of drugs that continue to interactwith hormonal contraception should beencouraged to consider a contraceptive methodthat is unaffected by the interacting drug (GoodPractice Point).

The contraceptive efficacy of hormonal contraception[with the exception of the progestogen-only injectable orthe levonorgestrel-releasing intrauterine system (LNG-IUS)] is reduced by liver enzyme-inducing drugs.Non-liver enzyme-inducing antibiotics are used relativelyfrequently. The enterohepatic circulation of EE may bereduced and therefore the efficacy of combinedcontraception may be reduced. Evidence for this will beprovided in the following sections.

Women using hormonal contraception should beadvised that some drugs could reduce contraceptiveefficacy. Women should be advised to seek advice aboutinteracting medication from their general practitioner, localfamily planning service, local pharmacy or the fpa (FamilyPlanning Association) in the first instance when a new drugis used.

The length of time a woman has to take a drug thatinteracts with her hormonal contraception may influenceher contraceptive choices. For example, a woman usingCOC and given a short course of non-liver enzyme-inducing antibiotic (e.g. treatment of urinary tractinfection) or a liver enzyme-inducing antibiotic (e.g.prophylaxis against meningitis) may continue with COCwith additional contraception, such as condoms, untilcontraceptive efficacy is restored. A woman using COC

Microsomal enzymes involved in metabolism ofcontraceptive hormones and other drugs are found in liverand intestinal mucosal cells. There are two types ofmicrosomal enzymes: Phase I enzymes (mixed-functionoxidases), which catalyse oxidation, reduction andhydrolysis; and Phase II enzymes, which catalyseglucuronidation, sulphation and acetylation.21–23

Cytochrome P-450 is the most important family ofenzymes in drug metabolism and CYP3A4 is the majorsubtype found in adult hepatocytes and intestinal mucosalcells. There is marked interindividual variation in theactivity of cytochrome P-450. Drugs that inhibit or inducecytochrome P-450 may affect concurrent medications.

If cytochrome P-450 is induced, the metabolism ofconcomitant drugs is increased, potentially reducing theclinical effect. Drugs known to be potent inducers of liverenzymes are listed in Table 1. Once started, these drugsmay induce liver enzymes within 2 days and the effects aremaximal within 1 week. After cessation, liver enzymesreturn to their normal functionality in 4 weeks.

If cytochrome P-450 is inhibited, the metabolism ofconcomitant drugs is decreased, potentially leading totoxicity. Inhibition of liver enzymes is less relevant forwomen using hormonal contraceptives, as toxicity with EE orprogestogens rarely occurs. Inhibition occurs as soon as theinhibitor appears in the liver and is maximal within 24 hours.

Excretion and the enterohepatic circulationAfter metabolism, conjugates of EE, unaltered EE andprogestogens are excreted into bile and subsequentlyreleased into the small intestine. Conjugates cannot beabsorbed from the small intestine.

In the large intestine, conjugates are broken down byhydrolytic enzymes released from colonic bacteria(clostridia, Bacteroides species, lactose-fermentingcoliforms and some staphylococci). Conjugates are brokendown into free (active) metabolites, which can bereabsorbed. These are eventually excreted in urine.


CEU GUIDANCE

ABSORPTION

Oral EE and progestogens areabsorbed from small intestine

METABOLISM(First-pass metabolism)

Conjugated with glucuronicacid and sulphate in LIVER

and intestinal mucosa

LIVER ENZYME INDUCTION

Quantitatively the most important hepaticenzyme is cytochrome P-450 mixed functionoxidase

Induction of cytochrome P-450 accelerates themetabolism of EE, but less known aboutprogestogens

Decrease in circulating concentration of EEand progestogens potentially reduces clinicaleffect

Liver enzyme-inducing drugs include:rifampicin, some anti-epileptics, St John’sWort, some antiretrovirals

ENTEROHEPATIC CIRCULATION

Wide inter- and intra-individual variation

Non-liver enzyme-inducing antibioticsreduce colonic bacteria thus reducing theconversion of inactive metabolites of EE toactive metabolites, which are reabsorbed

Sulphates and glucuronidesare excreted into bile and

into the small intestine

LARGE INTESTINE

Hydrolytic enzymes fromcolonic bacteria act on

conjugates of EE

ENTEROHEPATICCIRCULATION

Active metabolites ofEE are reabsorbed

from the large bowel

Reabsorbed andexcreted in urine

Progestogens notmetabolised in bowelto active metabolitesexcreted in faeces

Figure 1 Pharmacokinetics of ethinylestradiol (EE) and progestogens andinteraction with liver enzyme-inducing drugs and antibiotics

Anti-epileptic drugs that induce liver enzymes affecthormonal contraception by increasing the metabolism ofEE and progestogens. A randomised, controlled, double-blind, crossover study in women using a 30 µg EE COCshowed that oxcarbazepine reduced serum concentrationsof EE and progestogen.28 A case series of women usingphenobarbital found a significant decrease in EEconcentrations and increase in sex hormone-bindingglobulin.29 A prospective study of women using a 35 µg EECOC showed increased clearance of EE with topiramate.30

A pharmacokinetic study showed reduced EEconcentrations in women using a 50 µg EE COC andcarbamazepine and phenytoin.31

Some anti-epileptic drugs do not induce liver enzymesand therefore do not affect hormonal contraception. A small,randomised, double-blind, crossover trial of women using a30 µg EE COC and levetiracetam found no decrease in EEand progestogen. In addition, gonadotrophin andprogesterone concentrations were not increased.32 A small,randomised, double-blind, controlled trial in women using a30 µg EE COC showed that the use of vigabatrin did notappear to affect COC and liver enzymes were not induced.33

A prospective crossover study found that gabapentin did notaffect EE and progestogen concentrations.34 Case reportsprovide lower quality evidence.35,36 Sodium valproate35

and lamotrigine36 do not appear to affect thepharmacokinetics of oral contraceptives.

Progestogen-only contraception (POC). A small studyinvestigated nine women using a levonorgestrel (LNG)implant (Norplant®) with phenytoin and/or other anti-epileptic medication.37 Serum concentration of LNGdecreased and two pregnancies were reported.

An observational study of women on anti-epileptic liverenzyme-inducing drugs using the LNG-IUS showed onetrue failure giving a failure rate of 1.1 per 100 woman-years(95% CI 0.03–6.25).38 The dose of progestogen releasedinto the uterine cavity from the LNG-IUS is 1000 timesgreater than the uterine concentration seen followingprogestogen-only implants. Most of the contraceptive effectof the LNG-IUS is mediated via this direct release into theuterine cavity and is unaffected by metabolism in the liver.

Little evidence was identified for efficacy of POPs withanti-epileptic drugs.39 However, the efficacy is likely to bereduced as for other oral hormonal methods. Manufacturersof POPs do not recommend their continued used forwomen taking liver enzyme-inducing drugs.40–43

The SPC for depot medroxyprogesterone acetate(DMPA) suggests that the efficacy of DMPA is unaffectedby liver enzyme-inducing drugs.44

AntifungalsGriseofulvin is a liver enzyme-inducer that may decrease theserum concentration, and hence efficacy, of drugs alsometabolised by CYP3A4.45 The contraceptive efficacy ofhormonal contraception may be decreased and additionalcontraceptive protection is advised with griseofulvin.45 Casereports have described pregnancy in women using COC andgriseofulvin.46–49 The BNF suggests that griseofulvin andother antifungals (fluconazole, itraconazole, ketoconazole)reduce the efficacy of oestrogen and progestogencontraception.27 Nevertheless, these other antifungals do notappear to induce liver enzymes. The SPC for itraconazolestates there is no associated increase in metabolism of EE orprogestogen.50 Unintended pregnancy has been reported inwomen using ketoconazole51 and itraconazole.Breakthrough bleeding has been reported in COC userstaking itraconazole52,53 and ketoconazole54 but nopregnancies were identified in these studies. Evidence from

who requires long-term treatment of epilepsy ortuberculosis (with liver enzyme-inducing drugs) orfrequent courses of antibiotics (e.g. cystic fibrosis) shouldbe advised to consider another method of contraceptionthat is unaffected by concomitant drug use.

The underlying condition for which drugs are beingused can also affect contraceptive choices and should betaken into consideration when counselling women. Forexample, women with epilepsy using drugs which induceliver enzymes may choose to continue with an increaseddose of COC and condoms.25 For women with epilepsywho have associated memory problems, a contraceptivemethod that avoids daily pill taking and is unaffected bydrug use may be preferred. Full discussion of otherunderlying conditions is outside the scope of thisGuidance.

Which drugs may reduce the efficacy of hormonalcontraception?Drugs that induce liver enzymes (Table 1) are used to treata number of conditions and may reduce the efficacy ofhormonal contraception by increasing the metabolism ofEE and progestogens. The use of antibiotics that do notinduce liver enzymes may also reduce the efficacy of somehormonal contraceptives by altering colonic bacteria.Evidence on interactions between these two types of drugsand hormonal contraception are summarised in this section.

Liver enzyme-inducing drugs

5 Women should be informed that drugs whichinduce liver enzymes can reduce the efficacy ofcombined hormonal contraception, progestogen-only pills (POPs), and implants but do not appearto reduce the efficacy of progestogen-onlyinjectables or the LNG-IUS (Grade C).

Liver enzyme-inducing drugs increase the metabolism ofEE and progestogens.21–24,26 EE and progestogens have anarrow therapeutic range, and a decrease in bioavailabilitymay lead to decreased contraceptive efficacy. Table 1 liststhose drugs that induce liver enzymes and relevantassociated drugs that do not. Details on all liver enzyme-inducing drugs can be found in the BNF and readers arereferred to the most recent version for further details.27

Anti-epileptic drugsEpilepsy is a common disorder and clinicians are likely tosee women using anti-epileptic drugs and contraception.Some drugs used in the management of epilepsy are liverenzyme-inducers. Guidelines on the management ofwomen with epilepsy recommend use of drugs that aresafer in pregnancy.25 Monotherapy is advised wherepossible.25 When initiating or changing anti-epilepticdrugs, the implications for women using hormonalcontraception should be considered.25 Some anti-epilepticdrugs are also used in the management of conditions otherthan epilepsy, but the interactions are the same.

Combined hormonal contraception. Evidence concerningconcurrent use of anti-epileptic liver enzyme-inducingdrugs and hormonal contraception is variable and of poorquality. Most evidence relates to the use of COC. Studiesare inconsistent and measure different outcomes:pharmacokinetics of EE and progestogens, serumconcentration of EE and progestogens, ovulation assessedby ultrasound or progesterone levels, gonadotrophinconcentrations, breakthrough bleeding, and (rarely)pregnancy rates.


CEU GUIDANCE

small, randomised, crossover trials has been reassuring withregard to fluconazole.55,56 Oral fluconazole (150 mg) givenfor two or three consecutive cycles did not reduce serumconcentrations of progestogens and EE. A case seriesshowed lower serum EE concentrations in women usingCOC with fluconazole compared to COC alone.57 Noovulation or breakthrough bleeding was reported. Data forPOC and these antifungals are unavailable.

Liver enzyme-inducing antibioticsRifampicin and rifabutin (used in the treatment oftuberculosis or meningococcal meningitis prophylaxis) arepotent liver enzyme-inducers and case series haveidentified pregnancies in women using COC andrifampicin.58,59 The evidence to support rifampicin andrifabutin decreasing the efficacy of hormonal contraceptionis of poor quality.60–62

Combined hormonal contraception. Evidence for reducedefficacy of COC with rifampicin and rifabutin is from onenon-blind, randomised, controlled trial.60 This trial showedthat the pharmacokinetics of EE and the progestogen,norethisterone, were affected by both rifampicin andrifabutin (300 mg/day for 10 days). Serum concentrationsof EE and norethisterone were decreased; irregularbleeding was more common; and serum concentrations offollicule-stimulating hormone and luteinising hormonewere increased. Nevertheless, ovulation was not identified.Other small studies have shown evidence of ovulation inwomen using oral contraception and rifampicin.63–65

Progestogen-only contraception. Evidence for reducedefficacy of POPs with rifampicin is obtained from onesmall prospective study of nine women usingantitubercular treatment (rifampicin and ethambutol).66

The half-life of norethisterone increased following thecessation of rifampicin. There are no direct data on theefficacy of progestogen-only injectables with rifampicin orrifabutin but the SPCs suggest they are unaffected.44

Evidence on the efficacy of the LNG-IUS with rifampicinor rifabutin is poor. An observational study found a lowfailure rate for LNG-IUS users taking liver enzyme-inducing drugs (anti-epileptics and rifampicin).38 Evidencefor rifampicin or rifabutin decreasing the efficacy ofprogestogen-only implants is poor. A case report of awoman using an LNG implant with a liver enzyme-inducing drug (phenytoin) suggested decreased serumconcentrations of LNG during treatment associated withovulatory cycles.67

Although evidence showing that rifampicin andrifabutin decrease the efficacy of hormonal contraception isof poor quality, this does not imply no association. Thesedrugs are potent liver enzyme-inducers and therefore arelikely to reduce efficacy of hormonal contraception.

St John’s WortEvidence that St John’s Wort (SJW) decreases the efficacyof hormonal contraception is of poor quality.68 Aprospective cohort study in male volunteers showed anincrease in intestinal P-glycoprotein and intestinal andhepatic CYP34A.69 Evidence exists that other medicines(such as rifampicin) which induce these same microsomalenzymes are associated with contraceptive failure.59

Evidence exists that SJW increases the metabolism of othermedications.70–72 Case reports show that concentrations ofcyclosporin decrease with SJW use, leading to transplantedorgan rejection.73–76

A small, randomised, non-blinded, non-crossover trialfound no evidence of ovulation when a 20 µg COC was

used with SJW.77 This trial aimed to detect ovulation byboth serum progesterone and ultrasound scan. COCcompliance was assessed with electronic monitoring. Thestudy was limited by its small size and short duration (threecycles). Bleeding occurring on one or two occasionsbetween withdrawal bleeds was common in Cycle 1 for allgroups (treatment and control). In Cycles 2 and 3, bleedingsettled in most COC users. However, with the addition ofSJW, bleeding continued into Cycles 2 and 3. A furthernon-randomised study of women using COC (Ortho-Novum®) has shown an increase in breakthrough bleedingwith SJW use; but no evidence of ovulation.78 No datawere identified on SJW and progestogen-only methods.

Although there is a lack of good quality evidence toshow a reduction in efficacy of hormonal contraceptionwith SJW, this does not imply no association. As SJW is aliver enzyme-inducer, the Clinical Effectiveness Unit(CEU) upholds advice from the Committee on Safety ofMedicines (CSM) that women using hormonalcontraception should be cautious when using SJW. The useof SJW may need to be reconsidered and even stopped.70

The CEU suggests that advice for women using hormonalcontraception and SJW is as for other liver enzyme-inducing drugs.

Antiretroviral drugsSome antiretroviral drugs may reduce the efficacy ofhormonal contraception. Antiretroviral drugs such asprotease inhibitors (e.g. lopinavir, ritonavir) and non-nucleoside reverse transcriptase inhibitors (e.g. nevirapine)are metabolised by the CYP3A4 liver enzyme system andcan affect liver enzymes (Table 1). Women with HIV mayuse combination therapy where one or more drugs mayinduce liver enzymes. Readers should refer to the mostrecent BNF27 or to a useful website79 for furtherinformation on interactions with antiretrovirals. Fewstudies have been published which investigated thepharmacokinetics of EE and progestogens withantiretroviral drugs and none on the efficacy of hormonalcontraception. Serum concentrations of EE were reducedwhen women using a 50 µg EE COC were also takingritonavir but no pregnancies have been documented.80

Further information is given in Table 1.

Other drugsTacrolimus is an immunosuppressant used in theprevention of donor organ rejection.27 Tacrolimus isextensively metabolised via CYP3A4 isoenzyme.Tacrolimus has an inhibitory effect on CYP3A4-dependentmetabolism and therefore the concomitant use of drugsmetabolised via CYP3A4 may be affected by tacrolimus.81

The SPC on tacrolimus is unclear regarding hormonalcontraception.82 Nevertheless, the BNF suggests thatconcurrent use of tacrolimus and contraceptive hormonesmay increase levels of tacrolimus and decreasecontraceptive hormones, potentially reducing efficacy.27

No randomised trials were identified on use of tacrolimusand hormonal contraception.

Lansoprazole is a proton pump inhibitor. The SPCsuggests lansoprazole is a weak inducer of cytochromeP-450.83 The effects of lansoprazole on a 30 µg EE COCwere investigated in a randomised placebo-controlledstudy.84 No significant changes were identified in theserum concentration of EE and progestogens withlansoprazole use. The BNF does not suggest thatlansoprazole, or any other proton pump inhibitor, affectsthe efficacy of hormonal contraception.27

Bosentan is used in the management of pulmonaryhypertension.85 Bosentan can induce cytochrome P-450.


CEU GUIDANCE

No studies were identified which investigated hormonalcontraceptive use and bosentan. The SPC, however,suggests the avoidance of hormonal contraception inpatients on this drug as a main method of contraception, asdoes the BNF.27,85

Modafinil, used in the management of narcolepsy, hasbeen shown in vitro to induce liver enzymes.27,86 Its usemay decrease the efficacy of hormonal contraception.

Non-liver enzyme-inducing antibiotics

6 Women should be informed that non-liver enzyme-inducing antibiotics can reduce the efficacy ofcombined hormonal contraception but there is noreduction in the efficacy of progestogen-onlymethods (Grade C).

Antibiotics that do not induce liver enzymes may reducecontraceptive efficacy in other ways. Non-liver enzyme-inducing antibiotics temporarily decrease colonic bacteriaand thus inhibit the enterohepatic circulation of EE. Threeweeks after the initiation of antibiotics, gut flora haverecovered. Many studies investigating thepharmacokinetics of antibiotics and hormonalcontraceptives are of short duration and only investigateeffects in the initial weeks of antibiotic use. Antibiotics donot affect progestogens, as metabolites are inactive.

Pregnancies have been documented following the useof antibiotics.3,46–48,87–94 Nevertheless, case seriesreporting inadvertent pregnancy following antibiotic use donot identify causation and the potential for bias andconfounding is high. The influence of antibiotics on therisk of pregnancy for individual women is unclear, evenfrom data reported to the CSM.95 All trials investigatingthe use of antibiotics in women using hormonalcontraception are limited by their sample size, shortduration, inconsistent assessment of ovulation, and failureto eliminate bias. These limitations were highlighted in asystematic review on antibiotics and hormonalcontraception.96

Three small randomised trials suggest thatciprofloxacin and ofloxacin may not affect COC.97–99 Asmall, randomised, placebo-controlled, crossover study ofwomen using a variety of COCs found no differences inserum concentrations of gonadotrophins or oestradiol withconcomitant use of ciprofloxacin 500 mg twice daily.97 Nodata were collected on ovarian follicular activity or serumprogesterone to identify ovulation. A more recent Phase Idouble-blind placebo-controlled trial investigated theinteractions between ciprofloxacin and a 30 µg EE COC.98

No significant ovarian follicular activity was identified(low serum concentrations of oestradiol) and progesteronelevels indicated anovulation. A small, randomised, double-blind, crossover study investigated the effect of ofloxacin200 mg twice daily on ovulation in women using a 30 µgEE COC.99 No evidence of ovulation, by ultrasound scanand serum progesterone, was found.

Small prospective non-randomised studies have failedto show that ampicillin has any effect on gonadotrophinconcentration or progesterone in women using a COC with≥30 µg EE.100–102

A retrospective cohort study of women attendingdermatology clinics suggested an increased risk ofpregnancy in COC users with concomitant use ofantibiotics minocycline and cephalosporins.6 However, thepotential for bias in this study was high.

A small, non-randomised trial suggested tetracyclinedid not affect the pharmacokinetics of EE orprogestogens.103 One study has investigated the effect of

tetracycline on the pharmacokinetics of transdermalcombined contraception.104 No effect was seen on serumconcentrations of EE and progestogens.

A non-randomised trial investigated the use ofdoxycycline 100 mg twice daily for 7 days in women usinga 35 µg EE COC and did not identify any changes in thepharmacokinetics of EE or progestogens.105

It is clear that although many studies of variable qualityhave investigated the pharmacokinetics of EE andprogestogens in women using antibiotics, no studies havereliably investigated the efficacy of hormonalcontraception.

Antimalarial drugsNeither chloroquine nor primaquine have an effect onserum concentrations of EE or LNG in women using a 30µg EE COC.106 Doxycycline is commonly used in themanagement of malaria. Although the pharmacokinetics ofEE and progestogens appear unaffected by doxycycline,105

advice regarding additional contraceptive protection whiletaking the antibiotic and for 7 days after cessation is as forother antibiotics.

What advice should be given to women usinghormonal contraception and liver enzyme-inducingdrugs?Combined hormonal contraception

7 Women taking liver enzyme-inducing drugs whowish to use COC should choose a regimencontaining at least 50 µµg EE daily. Additionalcontraceptive protection, such as condoms, shouldbe used until 4 weeks after the liver enzyme-inducing drug has been stopped. Informationshould be given on the use of alternative methodsof contraception if liver enzyme-inducing drugs areto be used long term (Grade C).

8 Breakthrough bleeding does not necessarilyindicate low serum EE concentrations and risk ofovulation. Nevertheless, women using liverenzyme-inducing drugs with breakthroughbleeding may increase their dose of EE above 50 µµgdaily (Good Practice Point).

9 No evidence was identified that supports omittingor reducing the pill-free interval to reduce the riskof ovulation in women using liver enzyme-inducers(Good Practice Point).

10 Women using liver enzyme-inducing drugs mayuse a combined contraceptive patch withadditional contraceptive protection, such ascondoms, until 4 weeks after the liver enzyme-inducing drug has been stopped. Informationshould be given on the use of alternative methodsof contraception (Grade C).

11 Women using even short courses of rifampicin (forprophylaxis) should be advised to use additionalcontraception during the course and for 4 weeksafterwards (Grade C).

Based on limited evidence (but due to the consequencesshould an unintended pregnancy ensue), the CEUrecommends that women using combined hormonalcontraception should be advised that the efficacy of thesemethods might be reduced with liver enzyme-inducingdrugs.


CEU GUIDANCE

It has been established UK practice to start a COC withat least 50 µg EE daily for women taking liver enzyme-inducing drugs (Table 2).1,11,25,107,108 Despite lack ofevidence, in the light of the potentially serious sequelae ofunintended pregnancy, women are advised to use additionalcontraception, such as condoms, during use of a liverenzyme-inducing drug and for 28 days after the liverenzyme-inducer is stopped.25,11 The most commonly usedCOC containing 50 µg EE (Ovran®) was discontinued in2002. An alternative preparation containing 50 µgmestranol (Norinyl-1®) is available. Two studies haveprovided conflicting evidence on the bioequivalence of 50µg EE and mestranol.109,110 The CEU suggests analternative regimen using two low-dose COCs providing atotal daily dose of 50–60 µg EE (e.g. Loestrin 20® plusLoestrin 30®, or Marvelon® plus Mercilon®, or twoMicrogynon 30®). No trials have compared thebioavailibility to that of a single tablet. This use of COCs isoutside product licence.

Breakthrough bleeding has been suggested as anindicator of low serum hormone concentrations and risk ofovulation; however, no evidence was identified to supportthis. Nevertheless, it may be prudent to increase the dose ofEE if breakthrough bleeding occurs in women using liverenzyme-inducers. This may also control the breakthroughbleeding. Omitting or reducing the pill-free interval hasalso been advised for women using liver enzyme-inducing

drugs. No evidence was identified as to whether thisoutside licence use improves contraceptive efficacy.

Additional contraceptive protection, such as condoms,is advised when women using liver enzyme-inducers areusing combined contraceptive patches.111 Using more thanone patch at a time is not recommended.111 Considerationshould be given to the alternative use of progestogen-onlyinjectables, the LNG-IUS or an intrauterine device (IUD),which are unaffected by liver enzyme-inducing drugs.

Progestogen-only contraception

12 Women using liver enzyme-inducing drugs shouldbe advised that progestogen-only injectables areunaffected and can be continued with the usualinjection interval (Grade C).

13 Women using liver enzyme-inducing drugs in theshort term may choose to continue withprogestogen-only implants. Additionalcontraceptive protection, such as condoms, shouldbe used until 4 weeks after the liver enzyme-inducing drug has stopped. Information should begiven on the use of alternative contraception ifliver enzyme-inducing drugs are to be used longterm (Good Practice Point).


CEU GUIDANCE

Table 2 Advice regarding contraceptive use for women using liver enzyme-inducing drugs

Contraceptive method Advice for women using liver enzyme-inducing drugs

Combined hormonal contraception

Combined oral contraception (COC) Use a COC with at least 50 µg ethinylestradiol daily. This can be taken as a 30 µg COC plus a 20 µg COC or as two 30 µg COCs.

Additional contraceptive protection, such as condoms, is required when taking liver enzyme-inducers andfor 4 weeks after they are stopped.

Information should be given on the use of alternative methods which are unaffected by liver enzyme-inducers.

If additional contraception fails or is not used emergency contraception may be indicated.

Combined contraceptive patch Use one patch per week as for women not using liver enzyme-inducers.

Additional contraceptive protection, such as condoms, is required when taking liver enzyme-inducers andfor 4 weeks after they are stopped.

Information should be given on the use of alternative contraceptive methods if liver enzyme-inducers are to beused long term.

If additional contraception fails or is not used emergency contraception may be indicated.

Progestogen-only contraception

Progestogen-only pills (POPs) Advise alternative contraceptive methods.

Progestogen-only implants May continue with progestogen-only implants with additional contraceptive protection, such as condoms, whentaking liver enzyme-inducers and for 4 weeks after they are stopped.

Information should be given on the use of alternative contraceptive methods if liver enzyme-inducing drugs areto be used long term.

Progestogen-only injectables Progestogen-only injectables are unaffected by liver enzyme-inducers. Continue with the usual injection interval of12 weeks for depot medroxyprogesterone acetate and 8 weeks for norethisterone enanthate.

Progestogen-only emergency Take a total dose of 2.25 mg levonorgestrel as a single dose as soon as possible and within 72 hourscontraception of unprotected sex.

Consider the use of a copper IUD, which is unaffected.

Levonorgestrel-releasing No additional contraceptive protection required.intrauterine system

Non-hormonal methods

Copper-bearing intrauterine No additional contraceptive protection required.contraceptive device (IUD), barrier methods .

14 Women using POPs should be advised to consideralternative contraception if liver enzyme-inducingdrugs are used (Grade C).

15 Women can be advised that the LNG-IUS appearsto be unaffected by liver enzyme-inducing drugs(Grade B).

16 Women using liver enzyme-inducing drugs whorequire progestogen-only emergency contraception(POEC) should be advised: to take a total of 2.25mg LNG as a single dose as soon as possible andwithin 72 hours of unprotected sex; that this use isoutside the product licence; and about thealternative use of an IUD (Grade C).

Based on limited evidence (but due to the consequencesshould an unintended pregnancy ensue), the CEUrecommends that women using POPs and progestogen-onlyimplants should be advised that the efficacy of these methodsis reduced with liver enzyme-inducing drugs (Table 2).

Although no good evidence for reduced efficacy ofPOPs was identified, the SPCs for POPs advise againstcontinued use with liver enzyme-inducing drugs.40–42,112

The SPC for DMPA advises that the contraceptive efficacyis unaffected by liver enzyme-inducing drugs andinjections should be given at the usual 12-week interval.44

Similar advice (no need to reduce the 8-week injectioninterval) is given for norethisterone enanthate. The SPC forthe only progestogen-only implant (Implanon®) availablein the UK suggests additional contraceptive protectionwhile women are using a liver enzyme-inducing drug andfor 28 days after its cessation.113

Women not using liver enzyme-inducing drugs whorequire POEC are advise to take a single 1.5 mg dose ofLNG as soon as possible and within 72 hours ofunprotected intercourse.114 There are no data on the use ofPOEC by women using liver enzyme-inducing drugs.Clinical practice in the UK has been to increase the dose by50% (1.5 mg LNG at first presentation and 0.75 mg 12hours later).10 The most recent BNF, however, supportstaking 2.25 mg LNG as a single dose at first presentation.27

The CEU was unable to identify any new data to support asingle dose of 2.25 mg LNG for women taking liverenzyme-inducers. Some clinicians currently use a 1.5 mgLNG dose and repeat it 12 hours later. There is no evidenceon efficacy, compliance or side effects with any of theseregimens. The CEU advises that women should beinformed about the lack of data on efficacy of POEC whenusing liver enzyme-inducers and be offered an IUD as analternative. These regimens of POEC are outside theproduct licence.114 Clinicians may consider using theregimen that is most acceptable to an individual woman.Pharmacies should have a supply of a new single tablet1.5 mg dose of LNG (Levonelle One Step®) early in 2005.This may also be available for National Health Servicesupplies later in 2005. Women who are using liver enzyme-inducing drugs attending pharmacies for POEC should bereferred to a prescribing clinician.

What advice should be given to women usinghormonal contraception and non-liverenzyme-inducing antibiotics?

17 Women should be advised that pregnancies havebeen reported in COC users taking non-liverenzyme-inducing antibiotics, but the evidence doesnot generally support reduced COC efficacy andcausation (Grade B).

18 A COC user taking a short course (<3 weeks) of non-liver enzyme-inducing antibiotics should be advisedto use additional contraceptive protection, such ascondoms, during the treatment and for 7 days afterthe antibiotic has been stopped. If fewer than sevenactive pills are left in the pack after antibiotics havestopped, she should omit the pill-free interval (ordiscard any inactive pills) (Grade C).

19 A combined contraceptive patch user taking ashort course (<3 weeks) of non-liver enzyme-inducing antibiotics (except tetracycline) should beadvised to use additional contraceptive protection,such as condoms, during the treatment and for 7days after the antibiotic is stopped. If there are <7days remaining before her usual patch-free week,another patch should be applied when due forchanging and the patch-free week delayed by 7days (Grade C).

20 A woman who is an established user of non-liverenzyme-inducing antibiotics (≥3 weeks) does notrequire additional contraceptive protection whenstarting combined hormonal contraception unlessshe changes to a different antibiotic (Grade C).

21 Women should be informed that the efficacy ofprogestogen-only methods of contraception is notreduced by non-liver enzyme-inducing antibioticsand additional contraceptive protection is notrequired (Grade C).

22 Women using non-liver enzyme-inducingantibiotics (short- or long-term) who requirePOEC may be advised that the usual dose (1.5 mgwithin 72 hours of unprotected intercourse) isappropriate (Grade C).

It has been established practice in the UK to adviseadditional contraceptive protection for women usingcombined hormonal contraception with broad-spectrumantibiotics. Based on published low-quality evidence,antibiotics (broad-spectrum and narrow-spectrum) appearto have a limited effect on serum concentrations of EE andprogestogens. Some studies show no alteration inpharmacokinetics of EE and progestogens and continuedanovulation. No study has reliably investigated whethercombined hormonal contraceptive efficacy is reduced.Case reports have described pregnancies in women usinghormonal contraception and antibiotics. Although this doesnot confirm direct causation, the consequences of anunplanned pregnancy are such that the CEU suggests acautious approach when advising women. Additionalcontraceptive protection, such as condoms, is advised whenCOC users start or change any non-enzyme-inducingantibiotic (Table 3). Additional contraceptive protection isadvised during the antibiotic treatment and after theantibiotics have stopped until seven consecutive pills havebeen taken. If there are fewer than seven active pillsremaining in the pack, the pill-free interval should beomitted.11 Traditionally, such advice on additionalcontraceptive precautions has been restricted to broad-spectrum antibiotics. The CEU recognises that there isconfusion and lack of clarity regarding what constitutes abroad-spectrum antibiotic. For simplicity, on pragmaticgrounds, the CEU recommends that this advice is appliedto all non-liver enzyme-inducing antibiotics.

The efficacy of progestogen-only methods is notreduced with non-liver enzyme-inducing antibiotics.


CEU GUIDANCE

Which drugs increase serum levels of contraceptivehormones and what are the resulting clinical effects?Liver enzyme inhibitorsConcentrations of EE can be increased with use of the non-steroidal anti-inflammatory drugs, etoricoxib andvaldecoxib.27 The antilipid medication, rosuvastatin, canincrease serum concentrations of both EE andprogestogens. The clinical effect of increased EE andprogestogens is unclear. No evidence was identified ofincreased risk of venous thromboembolism with theseconcurrent medicines.

Which drugs can be increased or decreased byconcomitant use of contraceptive hormones?The bioactive pool of some drugs can be increased byconcomitant hormonal contraceptive use (e.g.corticosteroids, ciclosporin, theophylline and ropinirole)(Table 4). Depending on the nature and clinical effect of thedrug, potentially toxic effects can result.

An increase in corticosteroid concentrations has little ifany clinical effect.

Ciclosporin is a potent immunosuppressant.Neurotoxicity and nephrotoxicity have been demonstrated.Sex steroid hormones may increase plasma levels ofciclosporin leading to toxic effects.27

Theophylline is a bronchodilator with a narrow

therapeutic index and toxicity (arrhythmia andconvulsions) if serum concentrations increase.27 Sexsteroid hormones may increase plasma levels oftheophylline leading to toxic effects.27

Ropinirole is used in the management of Parkinson’sdisease and serum concentrations may increase withhormonal contraceptive use.27

The use of oestrogen and/or progestogen is suspected tohave an antagonistic effect on the clinical effect of someconcurrent medication (Table 4). Little evidence to supportthis was identified. It is suspected that the antihypertensiveeffects of a variety of medications, the anticoagulant effectof warfarin and phenidione, and the antidiabetic effects ofa variety of medications may be antagonised by use ofoestrogens.27 It must be remembered, however, that therisks of oestrogen-containing contraception for womenwith uncontrolled hypertension, diabetes with smallvessel disease, or with venous thromboembolic diseasebeing actively treated outweigh any benefits (WHOCategory 3)115 or pose an unacceptable health risk (WHOCategory 4). Therefore this clinical situation should not arise.

Miscellaneous drugsSibutramine hydrochloride (Reductil®), a centrally actingserotonin and noradrenaline reuptake inhibitor, is used asan anti-obesity agent. A small, single-blind, placebo-


CEU GUIDANCE

Table 3 Advice regarding contraceptive use for women using non-liver enzyme-inducing antibiotics

Contraceptive method Advice for women using non-liver enzyme-inducing antibiotics

Combined hormonal contraception

Combined oral contraception (COC) Established COC user

When taking a short course (<3 weeks) of any non-liver enzyme-inducing antibiotic additional contraceptiveprotection, such as condoms, is advised during the treatment and for 7 days after the antibiotic has been stopped.

If fewer than seven pills are left in the packet after antibiotics have stopped the pill-free interval should be omitted or any inactive pills discarded.

When any non-liver enzyme-inducing antibiotic is taken for ≥3 weeks additional contraceptive protection is no longerrequired. If a new antibiotic is prescribed, however, advice is as for short courses.

New COC user

When starting a COC if currently taking a short course (<3 weeks) of any non-liver enzyme-inducing antibioticadditional contraceptive protection, such as condoms, is advised during the treatment and for 7 days after the antibioticis stopped.

When starting a COC if any non-liver enzyme-inducing antibiotic has been taken for ≥3 weeks additionalcontraceptive protection is not required.

Combined contraceptive patch Established contraceptive patch users

If taking a short course (<3 weeks) of any non-liver enzyme-inducing antibiotic (except tetracycline) additional contraceptive protection, such as condoms, is advised during the treatment and for 7 days after the antibiotic isstopped.

If there are fewer than 7 days remaining before the usual patch-free week another patch should be applied (when duefor changing) so that the patch-free week is delayed by 7 days.

When any non-liver enzyme-inducing antibiotic is taken for ≥3 weeks additional contraceptive protection is no longerrequired. If a new antibiotic is prescribed, however, advice is as for short courses.

New patch user

When starting a contraceptive patch if currently taking a short course (<3 weeks) of any non-liver enzyme-inducingantibiotic additional contraceptive protection, such as condoms, is advised during the treatment and for 7 days afterthe antibiotic is stopped.

When starting a contraceptive patch if any non-liver enzyme-inducing antibiotic has been taken for ≥3 weeksadditional contraceptive protection is not required.

Progestogen-only methods Efficacy of progestogen-only methods (including emergency contraception) is not reduced by non-liver enzyme-inducing antibiotics.

Non-hormonal contraception Efficacy unaffected.

controlled trial showed that sibutramine did not impairefficacy of concomitantly administered COC.116

Terbinafine has been associated with breakthroughbleeding in women using POC.27 Despite extensivesearching, no further evidence on the use of terbinafine inwomen using hormonal contraception was identified. It isunclear if this drug induces liver enzymes; but it is unlikelyto do so.

Retinoids do not appear to induce liver enzymes and noreduction in hormonal contraceptive efficacy has beendemonstrated. A pharmacokinetic study showed areduction in serum concentrations of EE and norethisteronewhen women taking COC (Ortho-Novum) were also takingisotretinoin.117

References1 Elliman A. Interactions with hormonal contraception. J Fam Plann

Reprod Health Care 2000; 26: 109–111.2 Kovacs GT, Riddoch G, Duncombe P, Welberry L, Chick P, Weisberg

E, et al. Inadvertent pregnancies in oral contraceptive users. Med JAust 1989; 150: 549–551.

3 Young LK, Farquhar CM, McCowan LME, Roberts HE, Taylor J. Thecontraceptive practices of women seeking termination of pregnancyin an Auckland clinic. N Z Med J 1994; 107: 189–191.

4 Sparrow MJ. Pill method failures in women seeking abortion:fourteen years experience. N Z Med J 1998; 111: 386–388.

5 Sparrow MJ. Pregnancies in reliable pill takers. N Z Med J 1989; 102:879.

6 Helms SE, Bredle DL, Zajic J, Jarjoura D, Brodell RT, Krishnarao I.Oral contraceptive failure and antibiotics. J Am Acad Dermatol 1997;36: 705–710.

7 Stockley IH (ed.). Stockley’s Drug Interactions (6th edn). London,UK: Pharmaceutical Press, 2002.

8 Goldzieher JW. Selected aspects of the pharmacokinetics andmetabolism of ethinyl estrogens and their clinical implications. Am JObstet Gynecol 1990; 163(1 Pt 2): 318–322.

9 Faculty of Family Planning and Reproductive Health Care(FFPRHC). UK Selected Practice Recommendations forContraceptive Use. London, UK: FFPRHC, 2002.http://www.ffprhc.org.uk.

10 Faculty of Family Planning and Reproductive Health Care ClinicalEffectiveness Unit. FFPRHC Guidance (April 2003). Emergencycontraception. J Fam Plann Reprod Health Care 2003; 29(2): 9–16.

11 Faculty of Family Planning and Reproductive Health Care ClinicalEffectiveness Unit. FFPRHC Guidance (October 2003). Firstprescription of combined oral contraception. J Fam Plann ReprodHealth Care 2003; 29(4): 209–223.

12 US Food and Drug Administration (FDA). FDA approves Ovcon 35as the first chewable oral contraceptive tablet for women. FDA Talk

Paper T03-74. Rockville, MD: FDA, November 2003.http://http://www.fda.gov/bbs/topics/ANSWERS/2003/ANS01260.html.

13 Abrams LS, Skee D, Natarajan J, Wong FA. Pharmacokineticoverview of Ortho Evra‘/Evra‘. Fertil Steril 2002; 77: S3–S12.

14 Mulders TMT, Lounsbach GR. Ovulation inhibition with NuvaRing,the novel combined contraceptive vaginal ring. Gynecol Endocrinol2000; 14: 197.

15 Makarainen L, van Beek A, Tuomivaara L, Asplund B, Bennink HC.Ovarian function during the use of a single contraceptive implant:Implanon compared with Norplant. Fertil Steril 1998; 69: 714–721.

16 Bennink HJ. The pharmacokinetics and pharmacodynamics ofImplanon, a single-rod etonogestrel contraceptive implant. Eur JContracept Reprod Health Care 2000; 5: 12–20.

17 Oritz A, Hirol M, Stanczyk FZ, Goebelsmann U, Mishell DR. Serummedroxyprogesterone acetate (MPA) concentrations and ovarianfunction following intramuscular injection of depo-MPA. Journal ofClinical Endocrinology and Metabolism 1977; 44: 32-8.

18 Mishell DR Jr. Pharmacokinetics of depot medroxyprogesteroneacetate contraception. J Repod Med 1996; 41: 381–390.

19 Sang GW, Fotherby K, Howard G, Elder M, Bye P. Pharmacokineticsof norethisterone oenanthate in humans. Contraception 1981; 1:15–27.

20 Joshi JV, Hazari K, Shah RS, Chadha UC, Chitlange S, Gokral JS, etal. Serum progestogen and norethisterone levels following injectionof norethisterone enanthate in different sites and doses. Steroids 1989;53: 751–761.

21 Watkins PB. Drug metabolism by cytochromes P450 in the liver andsmall bowel. Gastroenterol Clin North Am 1992; 21: 511–526.

22 Spatzenegger M, Jaeger W. Clinical importance of hepaticcytochrome P450 in drug metabolism. Drug Metab Rev 1995; 27:397–417.

23 Meyer JM, Rodvold KA. Drug biotransformation by the cytochromeP-450 enzyme system. Infect Med 1996; 13: 452, 459, 463–464, 523.

24 Grimmer SFM, Back DJ, Orme ML’E, Cowie A, Gilmore I, Tjia J.The bioavailability of ethinyloestradiol and levonorgestrel in patientswith an ileostomy. Contraception 1986; 33: 51–59.

25 Scottish Intercollegiate Guidelines Network (SIGN) Guideline.Management of Women with Epilepsy. Edinburgh, UK: SIGN, 2003.

26 Akpoviroro J, Mangalam M, Kenkins N, Fotherby K. Binding ofcontraceptive steroids medroxyprogesterone acetate and ethinyloestradiol in the blood of various species. J Steroid Biochem 1981;14: 493–498.

27 British National Formulary, Vol. 48. London, UK: British MedicalAssociation and the Royal Pharmaceutical Association of GreatBritain, 2004. http://www.bnf.org.uk.

28 Fattore C, Cipolla G, Gatti G, Limido GL, Sturm Y, Bernasconi C, etal. Induction of ethinylestradiol and levonorgestrel metabolism byoxcarbazepine in healthy women. Epilepsia 1999; 40: 783–787.

29 Back DJ, Bates M, Bowden A, Breckenridge AM, Hall MJ, Jones H,et al. The interaction of phenobarbital and other anticonvulsants withoral contraceptive steroid therapy. Contraception 1980; 22: 495–503.

30 Rosenfeld WE, Doose DR, Walker SA, Nayak RK. Effect of


CEU GUIDANCE

Table 4 Drugs whose effects may be decreased or increased by use of hormonal contraception27

Type of drug Clinical effect

Decreased clinical effect

All antihypertensives Hypotensive effect may be antagonised by combined hormonal contraception.

Antidiabetics Hypoglycaemic effects maybe antagonised by combined hormonal contraception.

Anticoagulants Anticoagulant effect reduced by ethinylestradiol (EE) and progestogens.PhenidioneWarfarin

Antidepressants Antidepressant effects of tricyclics can be reduced by EE but side effects of tricyclics may be increased because serumTricyclics concentration is increased. No evidence identified.

Increased clinical effect

Immunosuppressants Serum concentrations of ciclosporin increased by EE and progestogens potentially leading to toxicityCiclosporin

.Corticosteroid Serum concentrations of corticosteroid increased by EE and progestogens but no significant clinical effect.

Bronchodilators Serum concentrations of theophylline increased by EE and progestogens potentially leading to toxicity.TheophyllineDopaminergics Serum concentrations of ropinirole increased by EE and progestogens but no significant clinical effect.Ropinirole

Potassium-sparing diuretics Potentially lead to hyperkalaemia when used with the progestogen drospirenone.

topiramate on the pharmacokinetics of an oral contraceptivecontaining norethindrone and ethinyl estradiol in patients withepilepsy. Epilepsia 1997;38: 317-323.

31 Crawford P, Chadwick DJ, Martin C, Tjia J, Back DJ, Orme M. Theinteraction of phenytoin and carbamazepine with combined oralcontraceptive steroids. Br J Clin Pharmacol 1990; 30: 892–896.

32 Ragueneau-Majlessi I, Levy RH, Janik F. Levetiracetam does notalter the pharmacokinetics of an oral contraceptive in healthy women.Epilepsia 2002; 43: 702.

33 Bartoli A, Gatti G, Cipolla G, Barzaghi N, Veliz G, Fattore C, et al. Adouble-blind, placebo-controlled study on the effect of vigabatrin onin vivo parameters of hepatic microsomal enzyme induction and onthe kinetics of steroid oral contraceptives in healthy femalevolunteers. Epilepsia 1997; 38: 702–707.

34 Eldon MA, Underwood BA, Randinitis EJ, Sedman AJ. Gabapentindoes not interact with a contraceptive regimen of norethindroneacetate and ethinyl estradiol. Neurology 1998; 50: 1146–1148.

35 Crawford P, Chadwick D, Cleland P, Tjia J, Cowie A, Back DJ, et al.The lack of effect of sodium valproate on the pharmacokinetics oforal contraceptive steroids. Contraception 1986; 33: 23–29.

36 Sabers A, Buchholt JM, Uldall P, Hansen EL. Lamotrigine plasmalevels reduced by oral contraceptives. Epilepsy Res 2001; 47:151–154.

37 Haukkamaa M. Contraception by Norplant subdermal capsules is notreliable in epileptic patients on anticonvulsant treatment.Contraception 1986; 33: 559–565.

38 Bounds W, Guillebaud J. Observational series on women using thecontraceptive Mirena® concurrently with anti-epileptic and otherenzyme-inducing drugs. J Fam Plann Reprod Health Care 2002; 28:78–80.

39 McCann MF, Potter LS. Progestin-only oral contraception: acomprehensive review. Contraception 1994; 50: S159–S188.

40 Wyeth Pharmaceuticals. Microval. 2003. http://www.medicines.org.41 Janssen-Cilag Ltd. Micronor Oral Contraceptive Tablets. 2001.

http://www.medicines.org.42 Pharmacia. Femulen Tablets. 2002. http://www.medicines.org.43 Organon Laboratories Ltd. Summary of Product Characteristics for

desogestrel-only pill (Cerazette). 1-6. 2004. http://www.medicines.org.

44 Pharmacia. Depot Medroxyprogesterone Acetate. 2004.http://www.medicines.org.

45 GlaxoSmithKline UK. Griseofulvin, Grisovin Tablets. 2004.http://www.medicines.org.

46 van DiJke CPH, Weber JCP. Interaction between oral contraceptivesand griseofulvin. BMJ 1984; 288: 1125–1216.

47 McDaniel PA, Caldroney RD. Oral contraceptives and griseofulvininteraction. Drug Intell Clin Pharm 1986; 20: 384.

48 Bollen M. Use of antibiotics when taking the oral contraceptive pill.Aust Fam Physician 1995; 24: 928–929.

49 Côte J. Interaction of griseofulvin and oral contraceptives. J Am AcadDermatol 1990; 22: 124–125.

50 Janssen-Cilag Ltd. Itraconazole, Sporanox Capsules. 2004.http://www.medicines.org.

51 Pillans PI. Pregnancy associated with a combined oral contraceptiveand itraconazole. N Z Med J 1993; 106: 436.

52 Meyboom RHB, van Puijenbroek EP, Vinks MHAM, Lastdrager CJ.Disturbance of withdrawal bleeding during concomitant use ofitraconazole and oral contraceptives. N Z Med J 1997; 110: 300.

53 van Puijenbroek EP, Egberts ACG, Meyboom RHB, Leufkens HGM.Signalling possible drug–drug interactions in a spontaneous reportingsystem: delay of withdrawal bleeding during concomitant use of oralcontraceptives and itraconazole. J Clin Pharmacol 1999; 47:689–693.

54 Kovacs I, Somos P, Hamori M. Examination of the potentialinteraction between ketoconazole (Nizoral) and oral contraceptiveswith special regard to products of low hormone content (Rigevidon,Anteovin). Ther Hung 1986; 34: 167–170.

55 Hilbert J, Messig M, Kuye O, Friedman H. Evaluation of aninteraction between fluconazole and oral contraceptives in healthywomen. Obstet Gynecol 2001; 98: 218–223.

56 Sinofsky FE, Pasquale SA. The effect of fluconazole on circulatingethinyl estradiol levels on women taking oral contraceptives. Am JObstet Gynecol 1998; 178: 300–304.

57 Devenport MH, Crook D, Wynn V, Lees LJ. Metabolic effects of low-dose fluconazole in healthy female users and non-users of oralcontraceptives. Br J Clin Pharmacol 1989; 27: 851–859.

58 Gupta KC, Ali MY. Failure of oral contraceptive with rifampicin. MedJ Zambia 1980; 15: 23.

59 Skolnick JL, Stoler BS, Katz DB, Anderson WH. Rifampicin, oralcontraceptives, and pregnancy. JAMA 1976; 236: 1382.

60 LeBel M, Masson E, Guilbert E, Colborn D, Paquet F, Allard S, et al.Effects of rifabutin and rifampicin on pharmacokinetics ofethinylestradiol and norethindrone. J Clin Pharmacol 1998; 38:1042–1050.

61 Barditch-Crovo P, Trapnell CB, Ette E, Zacur HA, Coresh J, Rocco

LE, et al. The effects of rifampin and rifabutin on thepharmacokinetics and pharmacodynamics of a combination oralcontraceptive. Clin Pharmacol Ther 1999; 65: 428–438.

62 Finch CK, Chrisman CR, Baciewicz AM, Self TH. Rifampin andrifabutin drug interactions. Arch Intern Med 2002; 162: 985–992.

63 Meyer BH, Muller FO, Wessels P. Lack of interaction betweenroxithromycin and an oral contraceptive. Eur J Pharmacol 1990; 193:1031–1032.

64 Meyer B, Müller F, Wessels P, Maree J. A model to detect interactionsbetween roxithromycin and oral contraceptives. Clin Pharmacol Ther1990; 47: 671–674.

65 Joshi JV, Joshi UM, Sankolli GM, Gupta K, Rao AP, Hazari K, et al.A study of the interaction of a low-dose contraceptive with anti-tubercular drugs. Contraception 1980; 21: 617–629.

66 Back DJ, Breckenridge AM, Crawford F, MacIver M, Orme MLE,Park BK, et al. The effect of rifampicin on norethisteronepharmacokinetics. Eur J Clin Pharmacol 1979; 15: 193–197.

67 Odlind V, Olsson SE. Enhanced metabolism of levonorgestrel duringphenytoin treatment in a woman with Norplant contraception.Contraception 1986; 33: 257–261.

68 Schwarz UI, Büschel B, Kirch W. Unwanted pregnancy on self-medication with St John’s Wort despite hormonal contraception. Br JClin Pharmacol 2003; 55: 112–113.

69 Durr D, Stieger B, Kullak-Ublick GA, Rentsch KM, Steinert HC,Meier PJ, et al. St John’s Wort induces intestinal P-glycoprotein/MDR1 and intestinal and hepatic CYP3A4. ClinPharmacol Ther 2000; 68: 598–604.

70 Chief Medical Officer (CMO). Fact Sheet for Health CareProfessionals. Interactions between St John’s Wort (Hypericumperforatum) preparations. CEM/CMO/2000/4. London, UK:Department of Health, 2000. http://open.gov.uk-mca/mcahome.htm.

71 Ruschitzka F, Meier PJ, Turina M, Lüscher TF, Noll G. Acute hearttransplant rejection due to Saint John’s wort. Lancet 2000; 355:548–549.

72 Piscitelli SC, Burstein AH, Chaitt D, Alfaro RM, Falloon J. Indinavirconcentrations and St John’s Wort. Lancet 2000; 355: 547.

73 Mai I, Kruger H, Budde K, Johne A, Brockmoller J, Neumayer HH,et al. Hazardous pharmacokinetic interaction of Saint John’s Wort(Hypericum perforatum) with the immunosuppressant cyclosporin.Int J Clin Pharmacol Ther 2000; 38: 500–502.

74 Bauer A, Stormer E, Johne A, Kruger H, Budde K, Neumayer H-N,et al. Alterations in cyclosporin A pharmacokinetics and metabolismduring treatment with St John’s Wort in renal transplant patients. Br JClin Pharmacol 2003; 55: 203–211.

75 Moschella C, Jaber BL. Interaction between cyclosporine andHypericum perforatum (St John’s wort) after organ transplantation.Am J Kidney Dis 2001; 38: 1105–1107.

76 Breidenbach T, Hoffmann MW, Becker T, Schlitt H, Klempnauer J.Drug interaction of St John’s wort with cyclosporin. Lancet 2000;355: 1912.

77 Pfrunder A, Schiesser M, Gerber S, Haschke M, Bitzer J, Drewe J.Interaction of St John’s Wort with low-dose oral contraceptivetherapy: a randomized controlled trial. Br J Clin Pharmacol 2003; 56:683–690.

78 Hall SD, Wang Z, Huang S, Hamman MA, Vasavada N, Adigun AQ,et al. The interaction between St John’s Wort and an oralcontraceptive. Clin Pharmacol Ther 2003; 74: 525–535.

79 Liverpool HIV Pharmacology Group (LHPG) and Department ofPharmacology and Therapeutics, University of Liverpool, Liverpool,UK. http://www.hiv-druginteractions.org.

80 Quellet D, Hsu A, Qian J, Locke CS, Eason CJ, Cavanaugh JH, et al.Effects of ritonavir on the pharmacokinetics of ethinyl oestradiol inhealthy female volunteers. Br J Clin Pharmacol 1998; 46: 111–116.

81 Lecointre K, Furlan V, Taburet AM. In vitro effects of tacrolimus onhuman cytochrome P450. Fundam Clin Pharmacol 2002; 16:455–460.

82 Fujisawa Limited. Tacrolimus Progaf Capsules. 2002.http://www.medicines.org.

83 Datapharm. Lansoprazole Summary of Product Characteristics.http://www.medicines.org.

84 Fuchs W, Sennewald R, Klotz U. Lansoprazole does not affect thebioavailability of oral contraceptives. Br J Clin Pharmacol 1994; 38:376–380.

85 Kobrin I. Bosentan therapy for pulmonary hypertension. 2001.http://www.medicines.org.

86 Robertson P, DeCory HH, Madan A, Parkinson A. In vitro inhibitionand induction of human hepatic cytochrome P450 enzymes bymodafinil. Drug Metab Dispos 2000; 28: 664–671.

87 Dossetor J. Drug interactions with oral contraceptives. BMJ 1984; 4:467–468.

88 Bacon JF, Shenfield GM. Pregnancy attributable to interactionbetween tetracycline and oral contraceptives. BMJ 1980; 280: 293.

89 Silber TJ. Apparent oral contraceptive failure associated withantibiotic administration. J Adolesc Health Care 1983; 4: 287–289.

90 Bainton R. Interaction between antibiotic therapy and contraceptive


CEU GUIDANCE

medication. Oral Surg Oral Med Oral Pathol Oral Radiol Endod1986; 61: 453–455.

91 Donley TG, Smith RF, Roy B. Reduced oral contraceptiveeffectiveness with concurrent antibiotic use: a protocol forprescribing antibiotics to women of childbearing age. Compendium1990; 11: 392–396.

92 DeSano EA, Hurley SC. Possible interactions of antihistamines andantibiotics with oral contraceptive effectiveness. Fertil Steril 1982;37: 853–854.

93 Bromham DR, Cartmill RSV. Knowledge and use of secondarycontraception among patients requesting termination of pregnancy.BMJ 1993; 306: 556–557.

94 London BM, Lookingbill DP. Frequency of pregnancy in acnepatients taking oral antibiotics and oral contraceptives. ArchDermatol 1994; 130: 392–393.

95 Back DJ, Grimmer SFM, Orme MLE, Proudlove C, Mann RD,Breckenridge AM. Evaluation of Committee on Safety of Medicinesyellow card reports on oral contraceptive–drug interactions withanticonvulsants and antibiotics. Br J Clin Pharmacol 1988; 25:527–532.

96 Weaver K, Glasier A. Interaction between broad-spectrum antibioticsand the combined oral contraceptive pill. A literature review.Contraception 1999; 59: 71–78.

97 Maggiolo F, Puricelli G, Dottorini M, Caprioli S, Bianchi W, Suter F.The effect of ciprofloxacin on oral contraceptive steroid treatments.Drugs Exp Clin Res 1991; 17: 451–454.

98 Scholten PC, Droppert RM, Zwinkels MGL, Moesker HL, Nauta JJP,Hoepelman IM. No interaction between ciprofloxacin and an oralcontraceptive. Antimicrobial Agents Chemother 1998; 42:3266–3268.

99 Csemiczky G, Alvendal C, Landgren BM. Risk for ovulation inwomen taking a low-dose oral contraceptive (Microgynon) whenreceiving antibacterial treatment with fluoroquinolone (ofloxacin).Adv Contracept 1996; 12: 101–109.

100 Friedman CI, Huneke AL, Kim MH, Powell J. The effect of ampicillinon oral contraceptive effectiveness. Obstet Gynecol 1980; 55: 33–37.

101 Back DJ, Breckenridge AM, MacIver M, Orme M, Rowe PH, StaigerCH, et al. The effects of ampicillin on oral contraceptive steroids inwomen. Br J Clin Pharmacol 1982; 14: 43–48.

102 Joshi JV, Joshi UM, Sankholi GM, Krishn U, Mandlekar A,Chowdhury V, et al. A study of interaction of low-dose combinationoral contraceptive with ampicillin and metronidazole. Contraception

1980; 22: 643–652.103 Murphy AA, Zacur HA, Charace P, Burkman RT. The effect of

tetracycline on level of oral contraceptives. Am J Obstet Gynecol1991; 164: 28–33.

104 Abrams LS, Skee DM, Natarajan J, Hutman W, Wong FA.Tetracycline HCL does not affect the pharmacokinetics of acontraceptive patch. Int J Gynaecol Obstet 2000; 70: 57–58.

105 Neely JL, Abate M, Swinker M, D’Angio R. The effect ofdoxycycline on serum levels of ethinyl estradiol, norethindrone, andendogenous progesterone. Obstet Gynecol 1991; 77: 416–420.

106 Back DJ, Breckenridge AM, Grimmer SFM, Orme MLE, Purba HS.Pharmacokinetics of oral contraceptive steroids following theadministration of the antimalarial drugs primaquine and chloroquine.Contraception 1984; 30: 289–295.

107 American Academy of Neurology. Practice parameter: managementissues for women with epilepsy. Report of the Quality StandardsSubcommittee of the American Journal of Neurology. Neurology1998; 51: 944–948.

108 Crawford P. Interactions between anti-epileptic drugs and hormonalcontraception. CNS Drugs 2002; 16: 263–272.

109 Goldzieher JW, Brody SA. Pharmacokinetics of ethinyl estradiol andmestranol. Am J Obstet Gynecol 1990; 163: S2114–S2119.

110 Kisicki J. Bioequivalence of Norethin and Ortho-Novum in healthyfemales. Adv Ther 1989; 6: 261–268.

111 Janssen-Cilag International N.V. Evra Transdermal Patch. 1-16. 2002.http://www.medicines.org.

112 Organon Laboratories Ltd. Cerazette. 8-7-2004.http://www.medicines.org.

113 Organon Laboratories Ltd. Implanon. 1-13. 2004.http://www.medicines.org.

114 Schering Health Care Ltd. New dose instructions for Levonelle-2.2004. http://www.medicines.org.

115 World Health Organization (WHO). Medical Eligibility Criteria forContraceptive Use (3rd edn). Geneva, Switzerland: WHO, 2004.

116 Back DJ, Fearn R, Oliver S, Stott DA, Wynne RD. Lack of effect ofsibutramine and the efficacy of oral contraceptive steroids in healthywomen. Int J Pharm Med 1997; 11: 71–76.

117 Hendrix CW, Jackson KA, Whitmore E, Guidos A, Kretzer R, LissCM, et al. The effect of isotretinoin on the pharmacokinetics andpharmacodynamics of ethinyl estradiol and norethindrone. ClinPharmacol Ther 2004; 75: 464–475.


CEU GUIDANCE

This Guidance was developed by the Clinical Effectiveness Unit (CEU) of the Faculty of Family Planning and ReproductiveHealth Care (FFPRHC): Dr Gillian Penney (Director), Dr Susan Brechin (Co-ordinator) and Ms Gillian Stephen (Research Assistant)in consultation with the Clinical Effectiveness Committee, which includes service user representation and a multidisciplinary expertgroup of health care professionals involved in family planning and reproductive health care. The multidisciplinary group comprised:Dr Andrea Brockmeyer (Subspecialty Trainee in Sexual and Reproductive Health Care, Abacus Clinics, Liverpool), Mr Craig Rore(Lead Pharmacist, Grampian Medicines Information Centre, Aberdeen Royal Infirmary, Aberdeen), Dr Alyson Elliman (LeadAssociate Specialist, Family Planning Service, Croydon Primary Care Trust/FFPRHC Clinical Standards Committee Member), DrJames McClay (Senior Lecturer in Clinical Pharmacology, Department of Medicine and Therapeutics, Aberdeen Royal Infirmary,Aberdeen), Ms Susan Stewart (Helpline and Information Manager, Epilepsy Scotland, Glasgow), Dr Kate Weaver (Staff Grade,Family Planning and Reproductive Health Service, Dean Terrace, Edinburgh). Written feedback was provided by: Professor DavidBack (Professor of Pharmacology, University of Liverpool, Liverpool), Ms Toni Belfield (Director of Information, fpa, London), MsSheena Bevan (Epilepsy Fieldworker, Links Resource Centre, Aberdeen), Dr Rachel Westwick (Career Grade Trainee in FamilyPlanning and Reproductive Health Care/Trainee Member of the FFPRHC Education Committee).

This guidance is also available online at www.ffprhc.uk Evidence tables are available on the FFPRHC website. Thesesummarise relevant published evidence on drug interactions with hormonal contraception, which was identified and appraised in thedevelopment of this Guidance. The clinical recommendations within this Guidance (i.e. the text appearing within the blue and redboxes) are based on evidence whenever possible.

Electronic searches were performed for: MEDLINE (CD Ovid version) (1990–2004), EMBASE (1990–2004), PubMed(1990–2004), The Cochrane Library (to November 2004) and the US National Guideline Clearing House. The searches wereperformed using relevant medical subject headings (MeSH), terms and text words. The Cochrane Library was searched for systematicreviews, meta-analyses and controlled trials relevant to drug interactions with hormonal contraception. Previously existing guidelinesfrom the FFPRHC, the Royal College of Obstetricians and Gynaecologists (RCOG), the World Health Organization (WHO) andreference lists of identified publications were also searched. Similar search strategies have been used in the development of othernational guidelines. Selected key publications were appraised according to standard methodological checklists before conclusionswere considered as evidence. Evidence was graded as above, using a scheme similar to that adopted by the RCOG and other guidelinedevelopment organisations.

Grades of Recommendations

A Evidence based on randomised controlled trials (RCTs)

B Evidence based on other robust experimental or observational studies

C Evidence is limited but the advice relies on expert opinion and has the endorsement of respected authorities

Good Practice Point where no evidence exists but where best practice is based on the clinical

Documents

Drug Interactions Hormonal